Author Topic: Low-fee, lower performance vs. higher-fee, higher performance?  (Read 7489 times)

agoraphone

  • 5 O'Clock Shadow
  • *
  • Posts: 12
Low-fee, lower performance vs. higher-fee, higher performance?
« on: December 18, 2013, 09:02:01 AM »
I'm just getting started reading this blog and thinking about reworking my 401k funds, and I'm lucky enough to have access to some Vanguard funds in my 401k, namely VINIX, which is an S&P 500 tracking fund, charging .04% fees. So I've made that my default fund to contribute to, instead of a target-date fund which charges more like .65%. Of all the funds I'm invested in, the one with the highest fees (APHMX) charges 1.08%, which I suppose isn't terrible, but obviously isn't great either. So as I think about moving the money out of that fund, and into VINIX, I look at what each fund has earned. VINIX has earned 27.43% YTD and APHMX has earned 32.32%. So even after subtracting APHMX's higher fees, it has come out 3.82% ahead of VINIX.

I understand the "you can't consistently beat the market" line of thinking, but it's really hard for me to move money into a worse-performing fund to avoid a small fee difference. Thoughts?

matchewed

  • Magnum Stache
  • ******
  • Posts: 4381
  • Location: CT
Re: Low-fee, lower performance vs. higher-fee, higher performance?
« Reply #1 on: December 18, 2013, 09:25:04 AM »
You may just need to adjust your frame of reference. You're basically reacting to either a known (the fees) or an unknown (future performance).

Given a 100k investment you can either pay $1080 for APHMX or you can pay $40 for Vanguard. That is the known. I know what I'd pick.

The unknown is the future performance, can't really talk about it because it is unknown. You can't guarantee that APHMX will always beat VINIX by 1.04%. But you can guarantee what fees you choose to pay.

The second part is asset allocation APHMX is different than VINIX, APHMX is a Mid Cap fund while VINIX is a Large Cap fund. How does this fit into your asset allocation?

KingCoin

  • Pencil Stache
  • ****
  • Posts: 783
  • Location: Manhattan
  • Achieved FI @ 30
Re: Low-fee, lower performance vs. higher-fee, higher performance?
« Reply #2 on: December 18, 2013, 09:53:14 AM »
In a rising market, smaller capitalization companies usually outperform larger cap companies due to their higher risk profile. If the market reverses, VINIX will likely outperform. So, you're seeing exactly what you'd expect to see.

matchewed is right. You have to figure out how midcap stocks fit into your portfolio as a whole. If there's no cheaper substitute, it might be worth paying the higher fee.


agoraphone

  • 5 O'Clock Shadow
  • *
  • Posts: 12
Re: Low-fee, lower performance vs. higher-fee, higher performance?
« Reply #3 on: December 18, 2013, 11:32:25 AM »
Thanks for the replies!
The points around large vs. mid are well taken, but what if both funds were large cap?
For instance, another one of my funds is TRLGX. It's large growth, with an even higher YTD of 38.93% and fees of .57%.

I realize past performance doesn't guarantee the future, but when one fund outperforms the other, year after year, it makes me wonder if it's worth paying a fee for that. The 5 year performance on TRLGX is 24.66% versus VINIX at 17.61%. 10 year is 9.49% versus  7.69%. I also realize we're comparing granny smith vs. red delicious here because one is large growth and the other is large balanced.  I'm just having a hard time drinking the Vanguard low-fee coolaid when there are other funds which appear to consistently outperform. Am I missing something?

matchewed

  • Magnum Stache
  • ******
  • Posts: 4381
  • Location: CT
Re: Low-fee, lower performance vs. higher-fee, higher performance?
« Reply #4 on: December 18, 2013, 11:53:19 AM »
Thanks for the replies!
The points around large vs. mid are well taken, but what if both funds were large cap?
For instance, another one of my funds is TRLGX. It's large growth, with an even higher YTD of 38.93% and fees of .57%.

I realize past performance doesn't guarantee the future, but when one fund outperforms the other, year after year, it makes me wonder if it's worth paying a fee for that. The 5 year performance on TRLGX is 24.66% versus VINIX at 17.61%. 10 year is 9.49% versus  7.69%. I also realize we're comparing granny smith vs. red delicious here because one is large growth and the other is large balanced.  I'm just having a hard time drinking the Vanguard low-fee coolaid when there are other funds which appear to consistently outperform. Am I missing something?

You've already answered your own question, one is large growth, one is large balanced.

That aside you will always be able to find some exceptions to the rule, someone will always be out on the edge of the bell curve, no index fund advocate will deny that. Can someone stay out there? Maybe but odds are against that given that an investing timeline isn't measured in even 15 year increments but in 60. Will future fund managers have the same savvy? Will whatever method they use remain effective? For every exception you note there will be hundreds which aren't exceptions. If you're willing to bet your exception is truly exceptional then invest in it.

CB

  • Stubble
  • **
  • Posts: 112
Re: Low-fee, lower performance vs. higher-fee, higher performance?
« Reply #5 on: December 18, 2013, 11:57:34 AM »
To me better five and ten year performance measures do not equate to "consistently outperform."  If it were me making the decision I'd want to look at a much longer time horizon.

Another Reader

  • Walrus Stache
  • *******
  • Posts: 5216
Re: Low-fee, lower performance vs. higher-fee, higher performance?
« Reply #6 on: December 18, 2013, 12:21:14 PM »
5 and 10 year performance are certainly worth considering.  YTD is meaningless.  Somewhere you should have access to an analysis of the available funds that includes a chart of the hypothetical growth of $10,000 over 10 years.  That can be very helpful - a picture is pretty darn convincing.  If a fund can consistently outperform the index it most closely resembles and competing funds net of expenses, it's worth considering.  However, funds should be compared against competitors with the same asset classes in them.  Large cap value to large cap value and so on.

You also have to consider your investment horizon for the account.  You can invest and forget in index funds for 30 years and have a pretty good chance of having a nice nest egg at the end.  No fuss, no muss.  I like to look at shorter terms and manage more actively.  Therefore I hold a number of funds that have outperformed their peers for 10 years.  I also point the fire hose at the market sectors that I expect to do well and not at sectors that I expect to weaken.  I sold off the bond funds 9 months to a year ago and haven't regretted doing that for a minute. 

If you think the market behaves randomly, then index funds will allow you to sleep well.  I don't believe that and I invest accordingly.

KingCoin

  • Pencil Stache
  • ****
  • Posts: 783
  • Location: Manhattan
  • Achieved FI @ 30
Re: Low-fee, lower performance vs. higher-fee, higher performance?
« Reply #7 on: December 18, 2013, 01:15:51 PM »
How long has this fund been an option? Sometimes an administrator will add high fliers and dump duds. This introduces a selection bias that induces people to pick higher fee funds when in fact these funds are just lucky or have a philosophy that's aligned well with the last 10yrs of stock performance. It's a worthwhile exercise to figure out why they're outperforming. What have they done differently than the general market that has given them an edge? Is the edge skill based or did they just make one or two fortuitous macro/sector calls?
« Last Edit: December 18, 2013, 03:10:43 PM by KingCoin »

mm31

  • Stubble
  • **
  • Posts: 169
Re: Low-fee, lower performance vs. higher-fee, higher performance?
« Reply #8 on: December 18, 2013, 01:36:04 PM »
Please remember that while higher returns are expected, higher fees are guaranteed. If we all knew when higher fees would justify higher returns, investing in the stock market wouldn't be so hard.

wtjbatman

  • Handlebar Stache
  • *****
  • Posts: 1310
  • Age: 36
  • Location: Missouri
Re: Low-fee, lower performance vs. higher-fee, higher performance?
« Reply #9 on: December 18, 2013, 03:38:40 PM »
Thanks for the replies!
The points around large vs. mid are well taken, but what if both funds were large cap?
For instance, another one of my funds is TRLGX. It's large growth, with an even higher YTD of 38.93% and fees of .57%.

I realize past performance doesn't guarantee the future, but when one fund outperforms the other, year after year, it makes me wonder if it's worth paying a fee for that. The 5 year performance on TRLGX is 24.66% versus VINIX at 17.61%. 10 year is 9.49% versus  7.69%. I also realize we're comparing granny smith vs. red delicious here because one is large growth and the other is large balanced.  I'm just having a hard time drinking the Vanguard low-fee coolaid when there are other funds which appear to consistently outperform. Am I missing something?

Saying this would get a person lynched on Bogleheads, but a fee of .57 is not a deal breaker IMO. A lot of people who look at and invest in ETFs and Funds beyond index funds would say anything under, oh, .75 is worth looking at.

Cooperd0g

  • 5 O'Clock Shadow
  • *
  • Posts: 67
Re: Low-fee, lower performance vs. higher-fee, higher performance?
« Reply #10 on: December 18, 2013, 07:44:14 PM »

Saying this would get a person lynched on Bogleheads ...

Ha ha yes it would. And that is because the amount of mutual funds that can beat the index over a 5 year span, never mind a 10 year span, is like less than 5% (I forget the actual number, but it is pretty bad especially for 10 year time line). OP, if you are so omniscient that you can choose one of those 5% funds then you should just go to Vegas and get rich overnight.

Your reading assignment for the next few days is to to read the entire Bogleheads Wiki:

http://www.bogleheads.org/wiki/Main_Page

Another Reader

  • Walrus Stache
  • *******
  • Posts: 5216
Re: Low-fee, lower performance vs. higher-fee, higher performance?
« Reply #11 on: December 18, 2013, 08:32:47 PM »
And the actively managed mutual funds that have consistently outperformed over a 10 year or longer period are fairly easy to find with some work.  Most actively managed mutual funds are junk, run by companies not in business to produce the best results for the customer.  The better companies also have less successful funds.  Weeding out the garbage takes some patience.  When you find the consistent producers, you have to watch for changes in fund management or objectives and keep an eye on the performance.  But the research can be very fruitful.

agoraphone

  • 5 O'Clock Shadow
  • *
  • Posts: 12
Re: Low-fee, lower performance vs. higher-fee, higher performance?
« Reply #12 on: December 19, 2013, 07:46:25 AM »
OP, if you are so omniscient that you can choose one of those 5% funds then you should just go to Vegas and get rich overnight.

It seems that someone at my company has done some pretty good choosing for me in terms of making good funds available. I can choose from a total of 28 funds: 12 stock funds, 12 blend funds and 4 bond funds. Of the stock funds, 9 of them have a higher 10-year performance than VINIX. I realize that's not the only measure, and future performance requires a crystal ball, but the 10-year marks seem like a good starting place at least.

Thanks for the thoughts everyone! Sounds like we've got some folks in the "stick with low-fee index funds" camp and some in the "it's possible to do a little better" camp. You've given me a lot to think about.

hoppy08520

  • Stubble
  • **
  • Posts: 101
Re: Low-fee, lower performance vs. higher-fee, higher performance?
« Reply #13 on: December 19, 2013, 08:31:58 PM »
OP, if you are so omniscient that you can choose one of those 5% funds then you should just go to Vegas and get rich overnight.

It seems that someone at my company has done some pretty good choosing for me in terms of making good funds available. I can choose from a total of 28 funds: 12 stock funds, 12 blend funds and 4 bond funds. Of the stock funds, 9 of them have a higher 10-year performance than VINIX. I realize that's not the only measure, and future performance requires a crystal ball, but the 10-year marks seem like a good starting place at least.

Thanks for the thoughts everyone! Sounds like we've got some folks in the "stick with low-fee index funds" camp and some in the "it's possible to do a little better" camp. You've given me a lot to think about.
Anyone can pick nine funds that beat the S&P 500 index in the past. Do you realize that the people who pick funds for 401(k) plans do this every year? They drop last year's losers and pick last years's winners. What matters is future returns. How do you know these are "good funds" going forward? Paradoxically a fund that performed well in the recent past might underperform in the future due to cyclical returns and reversion to the mean.

One last tip: Google "performance chasing".
« Last Edit: December 19, 2013, 08:36:24 PM by hoppy08520 »

dmn

  • Stubble
  • **
  • Posts: 101
  • Location: Switzerland
Re: Low-fee, lower performance vs. higher-fee, higher performance?
« Reply #14 on: December 21, 2013, 03:54:04 AM »
And the actively managed mutual funds that have consistently outperformed over a 10 year or longer period are fairly easy to find with some work.
Sure, but finding them tells you nothing about their future performance. Follow me through the following thought experiment:

Assume that mutual funds do not outperform index funds, but just give market returns on average (before fees). Let us therefore assume that 50% overperform and 50% underperform every year. Think of these funds as being managed by monkeys buying and selling stocks literally randomly.

Mutual fund companies offer many different funds. If you start with 1000 funds 10 years ago, then on average one of them will have (randomly!) outperformed over 10 years straight, every year! (Because 1/2^10 = 1/1024.) This fund will be in the news, investors will flock to this fund hoping for outperformance. But the fund is still lead by a monkey who buys and sells randomly, so its future returns are still only market average - and below average if you have to pay a management fee to the monkey. It is just survivorship bias: people look at this one fund and think "10 years of consistent success, there must be a superior strategy behind this", ignoring the 500 funds that had a 10-year underperformance before fees, because those funds are eventually closed and forgotten.

Empirical studies show that (correcting for risk, i.e. choice of asset class) the performance of mutual funds is generally very close to that of monkeys buying and selling randomly. If you correct for fees, it is worse. That's why it does not make sense to select funds by past performance: you are probably just choosing the luckiest monkey, and his future returns are just as random as that of all the other monkeys in the game.

Another Reader

  • Walrus Stache
  • *******
  • Posts: 5216
Re: Low-fee, lower performance vs. higher-fee, higher performance?
« Reply #15 on: December 21, 2013, 05:18:12 AM »
Your assumptions are wrong.  If markets and the businesses whose shares are bought and sold in the markets behaved in a random fashion. and if the better fund managers that actually act in their customers' interest were really a bunch of monkeys, your argument might hold water. 

wtjbatman

  • Handlebar Stache
  • *****
  • Posts: 1310
  • Age: 36
  • Location: Missouri
Re: Low-fee, lower performance vs. higher-fee, higher performance?
« Reply #16 on: December 21, 2013, 05:36:47 AM »
Your assumptions are wrong.  If markets and the businesses whose shares are bought and sold in the markets behaved in a random fashion. and if the better fund managers that actually act in their customers' interest were really a bunch of monkeys, your argument might hold water.

He claimed there are empirical studies that prove mutual fund performance is as chaotic as a monkey buying and selling stocks at random. I'd like to see that quantitative data and peer reviewed research.

dmn

  • Stubble
  • **
  • Posts: 101
  • Location: Switzerland
Re: Low-fee, lower performance vs. higher-fee, higher performance?
« Reply #17 on: December 21, 2013, 06:12:10 AM »
Quote from: Another Reader
Your assumptions are wrong.
My point was that my assumptions are consistent with some mutual funds outperforming 10 years in a row.

Therefore you cannot infer that a good past performance means there are better chances of future overperformance. If some managers have superior strategy, you see some funds overperforming over a long time. If managers have no strategy at all, you will still see some funds overperforming over a long time, just by chance.

Therefore, assessing whether active management can beat the market requires a statistical analysis, and it is a very bad idea to just look at the specific 10-year past performance of an individual fund to determine whether it is a good idea to invest your money actively at high fees or passively at low fees. The large-scale statistical analyses that I know of find that active fund managers underperform the market after fees, and that the past track record for any individual fund does not tell you anything about its future returns. (It is difficult to tell from the data if they outperform slightly before fees or if their performance is truly random.)


I'd like to see that quantitative data and peer reviewed research.
One article I remember is from Malkiel: "The Efficient Market Hypothesis and its Critics" (the PDF can be found by googling the title). Look for the section titled "The Performance of Professional Investors".